On the Temporal Stability of Analyte Recognition with an E-Nose Based on a Metal Oxide Sensor Array in Practical...

Area of Science:

• Sensor Technology
• Analytical Chemistry
• Data Science

Background:

• Electronic nose (e-nose) systems are valuable for odor analysis but suffer from functional instability.
• This instability, occurring over various timescales, limits their practical deployment in real-world scenarios.
• Metal oxide sensor arrays are a common e-nose technology facing these challenges.

Purpose of the Study:

• To address the functional instability of electronic nose units.
• To demonstrate a method for improving e-nose stability using a metal oxide sensor array.
• To enhance the reliability and marketability of e-nose devices.

Main Methods:

• Utilized an e-nose based on a metal oxide sensor array.
• Employed open-air and headspace sampling techniques for odor analysis.
• Applied linear discriminant analysis (LDA) for multisensor signal processing.
• Introduced and tested an 'add-training' procedure to manage instabilities.

Main Results:

• Identified ambient air composition changes as a primary cause of e-nose instability.
• The add-training procedure successfully compensated for temporal ambient changes.
• The add-training method effectively managed long-term drift in multisensor array properties.
• Demonstrated improved stability of the e-nose system.

Conclusions:

• The add-training procedure offers a practical solution for e-nose functional instability.
• This method enhances the robustness of e-nose devices for long-term applications.
• The findings improve the prospects for the commercialization and widespread adoption of e-noses.